Research Guidelines

1. You should become certified in the FMS before starting your research

2. The FMS is scored on an ordinal scale and the reliability has been established

3. We recommend using the original scoring criteria (including the Asymmetry Score) as well as the research scoring criteria. Using 2 video cameras to collect data will make this easier.

4. You will have to establish your own reliability. We can help you with this design if needed.

5. The research scoring criteria can be treated as a continuous variable.

6. The Asymmetry Score needs to be treated as a separate variable.

7. For Injury Prediction Research:

• Since injury is multi-factorial, attempt to collect as many relevant risk factors as possible (e.g. previous injury, height, weight, etc)

• The best model for injury prediction is a prospective cohort study design

• Use logistic regression analysis to control for potential confounding variables (e.g. playing status, number of years playing sport, etc.)

• Sample size mainly depends on the expected effect of the risk factor on injury risk. Inadequate sample size is one of the biggest reason for not showing an association between a certain risk factor and injury. In general, to detect moderate to strong associations, 20–50 injury cases are needed. To detect small to moderate associations you would need about 200 injured subjects (e.g. if you expect a 20% injury rate, you would need about 200 subjects to start the study uninjured if you expect a strong correlation between the risk factor you are studying and injury. For a small association, you would need to start with 1000 athletes). Conduct a power analysis BEFORE beginning a study.

• Establish the reliability of the risk factor you are measuring

• Use a standardized time loss injury definition

• Use a systematized injury data collection system (e.g. Canadian Intercollegiate Sport Injury Registry) o (Adapted from Bahr 2003)

8. For intervention and/or correlation research:

• Utilize a standardized intervention

• Track compliance with intervention

• Blind measures to pre/post FMS (video)

• Use standard fitness tests which have been previously reported and blind when possible

• Use a self-reported measure such as global rate of change as part of post intervention questionnaire:

“How much better do you feel and in your performance after completing this program?” -7, -6 ………….-2, -1, 0, 1 ,2, …………6, 7. (This will allow for a dichotomized outcome variable)